Special Issue on Model-Based Testing

With the ever-increasing penetration of software-intensive systems into technical, business and social areas, not only the requirements on system functionality and features, but also the requirements on system quality and reliability are increasing. With the increasing requirements, the complexity of such software-intensive systems is growing – this combined with ever shortened development times has led to major problems with the classical (non-model-based) way of developing and testing such software. In order to remain competitive, an early and continuous consideration and assurance of system quality and reliability becomes an asset of ever-increasing importance in industrial software development. For high-quality software-intensive systems , it is known that testing takes 40-60% of the total development costs and that poorly tested software can result in product failure or even in catastrophic cases and losses of human life. Thus, it makes sense to develop and deploy new testing methods and tools to cope with the growing requirements, complexity and testing costs. Lately, an increased popularity of model-based approaches has been observed not only in the academic world but also in industrial environments. This increased interest can be traced back to the complexity and quality issues mentioned above, but it is also driven by the improved graphical support of modelling languages, the potential for testing automation , and improved tooling. This increase in popularity is also reflected in the software and system testing field, where the relatively old paradigm of model-based testing (MBT) has become better enabled and supported by the technological developments. The main idea behind model-based testing is the use of abstract specifications of the system under test (SUT) and/or of its environment, which are used for checking the implementation of the system against its specification. This is done by using these specifications for automatically designing , deriving, validating, and optimizing tests. Model-based testing refers to software testing where test cases are derived in whole or in part from a model that describes selected, often structural, functional, sometimes non-functional aspects of the system under test and/ or its environment. In model-based testing, the efforts traditionally put in eliciting test requirements and in